One automated notification template produced 97.5% of 303,314 “stealer” posts collected between January and April 2026. Source: Brandefense Cybercrime Report 2026.
Underground posts from 2026 show ransomware claims concentrated in a few groups, keyword counts inflated by bots, and “AI” and “zero-day” used as sales labels
WILMINGTON, DE, UNITED STATES, November 17, 2026 /EINPresswire.com/ — Brandefense, an AI-driven threat intelligence platform, today announced its Cybercrime Report 2026, now free to read at https://brandefense.io/cybercrime-report-2026/. The report analyzes what was posted, advertised and claimed in underground channels, mostly on Telegram, between January 1 and October 27, 2026. It draws on Brandefense Blackhole, the company’s cybercrime database. Its eight chapters show how bots, relays, copied offers and sales labels can change the apparent scale of underground activity, and what each count can and cannot show.
A few groups post most ransomware claims
• 9,203 leak-site claims came from 177 groups.• 33 of those groups made 64.7% of the claims.• Two groups, Qilin and The Gentlemen, made 23.4%.• In six selected affiliate-program ads, the partner keeps the larger share: 70% to 95%, with three ads at 90%.
Bots, repeated pitches and news retellings inflate keyword counts
• 303,314 posts used the word “stealer” between January and April. One automated notification template produced 97.5% of them.• A narrowly defined cash-withdrawal recruitment pitch appears in 831 of 2,448 retained financial-account posts, under 132 source names. All 831 texts differ, but three recruitment phrases recur in every one.• An estimated 63% of posts that mention “data breach” are news retellings, not leak posts.
“AI” and “zero-day” are often sales labels
• Brandefense read 607 texts from searches for seven AI-branded tools, including WormGPT; 168 of them were offers. The names mostly label a site, bot, app or prompt pack, and the ads rarely show which model runs underneath.• In a sample of 700 classified posts, 93 exploit seller posts use the word “0-day” and none names a CVE identifier. Only 37 were coded as claims of an unnamed exploit.• 56 of 193 seller posts in the same sample sell a mass-exploitation tool or the web shells it produces.• Web software is the largest stated target in exploit offers and requests, at 43%.
One hacktivist group’s claims follow a weekly pattern
• The group NoName057(16) made 578 DDoS claims listing 2,920 target entries.• Its claims follow weekly country themes: 42 of the 48 single-country runs of five or more claims began on a Monday.• Outage wording appears in the header of only 13% of the claims.
“Security teams hear every day that the underground is growing,” said Caner Köroğlu, CEO of Brandefense. “Our data shows something more specific: ransomware claims concentrated in a small number of groups, keyword counts inflated by automation, and a lot of sales language. This report separates the numbers that deserve attention from the noise, and each chapter publishes its method and limits so that readers can see how every figure was counted.”
The figures count posts, claims or coded samples from the Brandefense Blackhole collection, most of them from Telegram. They are not confirmed incidents, victims or market totals, and advertised prices and terms are sellers’ claims. AI agents read, labelled and analyzed the collected posts, and Brandefense analysts approved the findings.
About Brandefense
Brandefense is an AI-driven threat intelligence platform that unifies Cyber Threat Intelligence, Digital Risk Protection and External Attack Surface Management. Brandefense is a member of the Cyber Threat Alliance, and Brandefense CTI is an Associate member of FIRST. Learn more at https://brandefense.io/.Brandefense Research TeamBrandefensereports@brandefense.ioVisit us on social media:LinkedIn
Legal Disclaimer:
EIN Presswire provides this news content “as is” without warranty of any kind. We do not accept any responsibility or liabilityfor the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in thisarticle. If you have any complaints or copyright issues related to this article, kindly contact the author above.
Click Here For The Original Source.
——————————————————–
